MODELING OF SELECTION PROCESSES OF CLOUD SYSTEMS PARAMETERS PROVIDING THEIR STABILITY IN ACCORDANCE WITH RELIABILITY AND SAFETY

• Published in: Nauchno-tekhnicheskiĭ vestnik informat͡s︡ionnykh tekhnologiĭ, mekhaniki i optiki Vol. 18; no. 4; p. 654
• Main Authors: Bezzateev, S V, Elina, T N, V. A. Myl’nikov
• Format: Journal Article
• Language: Russian
• Published: Saint Petersburg St. Petersburg National Research University of Information Technologies, Mechanics and Optics 01.07.2018
• Subjects: Artificial neural networks; Cloud computing; Computer architecture; Fuzzy logic; Fuzzy systems; Mathematical models; Process parameters; Reliability analysis; Reliability aspects; Safety; User satisfaction
• ISSN: 2226-1494; 2500-0373
• DOI: 10.17586/2226-1494-2018-18-4-654-662

We have carried out the analysis of commercial and free software for support and organization of cloud computing, outlined the advantages and disadvantages of existing methods for reliability and security improvement of computing systems. Most of the existing systems do not take into account a number of factors that affect the safety, reliability and performance of calculations, the complexity of adaptation to changing requirements and environmental conditions. The work objective is formulated consisting in selection of cloud computing system architecture that provides maximum satisfaction of requests with different priority level, coming both from users and from services of the system itself. To solve this problem we propose a method of the system configuring for cloud services based on the model of a neuro-fuzzy system. The method gives the possibility to increase the productivity of users' requests providing the reliability and security of the processed information in special-purpose and dual-use systems. The architecture of the neuro-fuzzy network is developed, its input and output parameters are determined. Applying the proposed models, the configuration of a cloud information system designed to solve certain groups of tasks is carried out as an example. The decision result was a distribution matrix of system resources for serving of different task groups.